Adds groq inference adapter

llama_stack/templates/groq/__init__.py

@@ -0,0 +1,7 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .groq import get_distribution_template  # noqa: F401

llama_stack/templates/groq/build.yaml

@@ -0,0 +1,19 @@
+version: '2'
+name: groq
+distribution_spec:
+  description: Use Groq for running LLM inference
+  docker_image: null
+  providers:
+    inference:
+    - remote::groq
+    memory:
+    - inline::faiss
+    - remote::chromadb
+    - remote::pgvector
+    safety:
+    - inline::llama-guard
+    agents:
+    - inline::meta-reference
+    telemetry:
+    - inline::meta-reference
+image_type: conda

llama_stack/templates/groq/doc_template.md

@@ -0,0 +1,66 @@
+---
+orphan: true
+---
+# Groq Distribution
+
+```{toctree}
+:maxdepth: 2
+:hidden:
+
+self
+```
+
+The `llamastack/distribution-{{ name }}` distribution consists of the following provider configurations.
+
+{{ providers_table }}
+
+{% if run_config_env_vars %}
+### Environment Variables
+
+The following environment variables can be configured:
+
+{% for var, (default_value, description) in run_config_env_vars.items() %}
+- `{{ var }}`: {{ description }} (default: `{{ default_value }}`)
+{% endfor %}
+{% endif %}
+
+{% if default_models %}
+### Models
+
+The following models are available by default:
+
+{% for model in default_models %}
+- `{{ model.model_id }}`
+{% endfor %}
+{% endif %}
+
+### Prerequisite: API Keys
+
+Make sure you have access to a Groq API Key. You can get one by signing up at [console.groq.com](https://console.groq.com).
+
+## Running Llama Stack with Groq
+
+You can do this via Conda (build code) or Docker which has a pre-built image.
+
+### Via Docker
+
+This method allows you to get started quickly without having to build the distribution code.
+
+```bash
+LLAMA_STACK_PORT=5001
+docker run \
+  -it \
+  -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
+  llamastack/distribution-{{ name }} \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```
+
+### Via Conda
+
+```bash
+llama stack build --template {{ name }} --image-type conda
+llama stack run ./run.yaml \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```

llama_stack/templates/groq/groq.py

@@ -0,0 +1,76 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from pathlib import Path
+
+from llama_models.sku_list import all_registered_models
+
+from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput
+from llama_stack.providers.remote.inference.groq import GroqImplConfig
+from llama_stack.providers.remote.inference.groq.groq import MODEL_ALIASES
+from llama_stack.templates.template import (
+    DistributionTemplate,
+    RunConfigSettings,
+)
+
+
+def get_distribution_template() -> DistributionTemplate:
+    providers = {
+        "inference": ["remote::groq"],
+        "memory": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
+        "safety": ["inline::llama-guard"],
+        "agents": ["inline::meta-reference"],
+        "telemetry": ["inline::meta-reference"],
+    }
+
+    inference_provider = Provider(
+        provider_id="groq",
+        provider_type="remote::groq",
+        config=GroqImplConfig.sample_run_config(),
+    )
+
+    core_model_to_hf_repo = {
+        m.descriptor(): m.huggingface_repo for m in all_registered_models()
+    }
+
+    default_models = [
+        ModelInput(
+            model_id=core_model_to_hf_repo[m.llama_model],
+            provider_model_id=m.provider_model_id,
+        )
+        for m in MODEL_ALIASES
+    ]
+
+    return DistributionTemplate(
+        name="groq",
+        distro_type="self_hosted",
+        description="Use Groq for running LLM inference",
+        docker_image=None,
+        template_path=Path(__file__).parent / "doc_template.md",
+        providers=providers,
+        default_models=default_models,
+        run_configs={
+            "run.yaml": RunConfigSettings(
+                provider_overrides={
+                    "inference": [inference_provider],
+                },
+                default_models=default_models,
+                default_shields=[
+                    ShieldInput(shield_id="meta-llama/Llama-Guard-3-8B")
+                ],
+            ),
+        },
+        run_config_env_vars={
+            "LLAMASTACK_PORT": (
+                "5001",
+                "Port for the Llama Stack distribution server",
+            ),
+            "GROQ_API_KEY": (
+                "",
+                "Groq API Key for authentication",
+            ),
+        },
+    )

llama_stack/templates/groq/run.yaml

@@ -0,0 +1,95 @@
+version: '2'
+image_name: groq
+docker_image: null
+conda_env: groq
+apis:
+- agents
+- inference
+- memory
+- safety
+- telemetry
+providers:
+  inference:
+  - provider_id: groq
+    provider_type: remote::groq
+    config:
+      url: https://api.groq.com/openai/v1
+      api_key: ${env.GROQ_API_KEY}
+  memory:
+  - provider_id: faiss
+    provider_type: inline::faiss
+    config:
+      kvstore:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/faiss_store.db
+  safety:
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
+    config: {}
+  agents:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      persistence_store:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/agents_store.db
+  telemetry:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+metadata_store:
+  namespace: null
+  type: sqlite
+  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/registry.db
+models:
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-1B-Instruct
+  provider_id: null
+  provider_model_id: llama-3.2-1b-preview
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-3B-Instruct
+  provider_id: null
+  provider_model_id: llama-3.2-3b-preview
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
+  provider_id: null
+  provider_model_id: llama-3.2-11b-vision-preview
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
+  provider_id: null
+  provider_model_id: llama-3.2-90b-vision-preview
+- metadata: {}
+  model_id: meta-llama/Llama-3.1-8B-Instruct
+  provider_id: null
+  provider_model_id: llama-3.1-8b-instant
+- metadata: {}
+  model_id: meta-llama/Llama-3.1-70B-Instruct
+  provider_id: null
+  provider_model_id: llama-3.1-70b-versatile
+- metadata: {}
+  model_id: meta-llama/Llama-3-8B-Instruct
+  provider_id: null
+  provider_model_id: llama3-8b-8192
+- metadata: {}
+  model_id: meta-llama/Llama-3-70B-Instruct
+  provider_id: null
+  provider_model_id: llama3-70b-8192
+- metadata: {}
+  model_id: meta-llama/Llama-3-8B-Instruct
+  provider_id: null
+  provider_model_id: llama3-groq-8b-8192-tool-use-preview
+- metadata: {}
+  model_id: meta-llama/Llama-3-70B-Instruct
+  provider_id: null
+  provider_model_id: llama3-groq-70b-8192-tool-use-preview
+- metadata: {}
+  model_id: meta-llama/Llama-Guard-3-8B
+  provider_id: null
+  provider_model_id: llama-guard-3-8b
+shields: []
+memory_banks: []
+datasets: []
+scoring_fns: []
+eval_tasks: []